Re: Distributed erlang question

>
> erl -setcookie bob -name [email protected]
>

The above translates to: start an Erlang VM with cookie bob and a name of
[email protected]

erl -setcookie bob -name [email protected] -remsh [email protected]


And this translates to: start an Erlang VM with cookie bob and a name of
[email protected] and then open a remote shell to node with name
[email protected]. Your [email protected] node will connect to epmd on
port 4369 running on 192.168.86.35 and check if there is any node running
on that host with name [email protected]. This will fail because the
Erlang node running on that host has a name of [email protected].

In this working example we form a three-node cluster:
https://github.com/rabbitmq/rabbitmq-prometheus/blob/master/docker/docker-compose-overview.yml
. All nodes use rabbit@hostname where hostname is resolved by Docker.

There is a quick start if it helps:
https://www.rabbitmq.com/prometheus.html#quick-start

On Tue, Sep 22, 2020 at 8:07 PM Matthew Evans <[email protected]>
wrote:

> Hi,
>
> We want to start an Erlang instance within a docker container. What we
> want to do is start the Erlang VM like:
>
> erl -setcookie bob -name [email protected]
>
> Since we may not have the interfaces up at that time we provide the lo
> address.
>
> When the interfaces are running we want to be able connect to that VM from
> a remote node with the actual address:
>
> erl -setcookie bob -name [email protected] -remsh [email protected]
>
> This fails...
>
> I know we can probably do this programmatically instead by starting the VM
> with no distribution and call net_kernel:start/1when the interface is
> ready. Question: is there a better way?
>
> Thanks
>
>